# Thermohaline circulation The deep ocean turns over on a scale of centuries, driven by [[Density]] differences that come from temperature and salinity — thermo and haline. Cold salty water formed at high latitudes sinks, spreads along the bottom, and returns to the surface elsewhere by slow upward mixing. It is the slowest and largest circuit on this spine, and the ocean's share of [[Heat_transfer|heat transport]]. ## Microsim — p5.js Sidecar: `Thermohaline_circulation.p5.js` · route leaf: `microsim/p5js/Thermohaline_circulation__20260910T0400Z`. Built to the Betterfire Standard v0 (single `ARTICLE` constant, HUD title and Wikitube URL, control hints, parameter readout and equation). | Control | Does | |---------|------| | `cooling slider` | heat loss at high latitude | | `freshwater slider` | freshwater input there | | `[r]` | reset the overturning | *What to watch:* Cool the northern box and it sinks and the loop spins up. Now freshen it instead — the loop stalls and can reverse, even with the cooling unchanged. ## Sinking is local, rising is not Deep [[Water|water]] forms in only a few small places — chiefly the Nordic and Labrador seas and around Antarctica — where surface water is cold enough and salty enough to sink — a [[Density]] threshold, not a location. Return flow is the opposite: diffuse, spread over the whole ocean, and driven by mixing that lifts dense water back up against gravity. That mixing is supplied by tides and winds working against stratification, so the circulation's speed is set as much by [[Turbulence]] and [[Tide|tidal]] mixing as by the density contrast that starts it. ## Salt and heat pull in different directions Cooling makes water denser; freshening makes it lighter. Add enough fresh [[Water]] at high latitudes — from ice melt or rainfall — and sinking can weaken even while cooling continues, a [[Feedback|feedback]] with no obvious floor. Because the circulation also carries heat poleward, that produces a feedback the climate record suggests has switched abruptly in the past. The same competition between two density-setting variables appears at small scale as double diffusion, and it is what makes the [[Thermocline]] and its salinity twin persist rather than mixing away. **Reads with:** *Introduction to Physical Oceanography (Robert H. Stewart, 2008)* — [OTL record](https://open.umn.edu/opentextbooks/textbooks/introduction-to-physical-oceanography) · [download](https://github.com/introocean/introocean-en/releases/tag/v20200229) · CC BY-NC-SA. Section 7 of the [[PORTAL_Thury_Hydrodynamics_Apex_Spine|Apex Spine]] book shelf. **On the spine:** [[Ocean_current]] · [[Thermocline]] · [[Density]] · [[Upwelling]] · [[WT!Thury_Hydrodynamics_Compendium]]. <!-- COMPENDIUMLINK:BEGIN g19 — generated from _registry/plans/THURY_COMPENDIUM_SECTIONS.md; do not hand-edit inside --> *Linked from the [[WT!Thury_Hydrodynamics_Compendium|Thury Hydrodynamics Compendium]], section 18, Geophysical fluid dynamics.* <!-- COMPENDIUMLINK:END --> ## Wikipedia : Wikitube **Strict pair:** [Wikipedia](https://en.wikipedia.org/wiki/Thermohaline_circulation) : [Wikitube](https://en.wikitube.io/wiki/Thermohaline_circulation) ## Previous hub tags Hubs: `Life_Physics`, `Systems`. Portals: [[PORTAL_Thury_Hydrodynamics_Apex_Spine]], [[PORTAL_WT!Thury_Hydrodynamics_Compendium]], [[PORTAL_Physics]]. --- *Book-section wave · 2026-09-10 · article + p5 microsim shipped together.*